41 8 MANIPULATION AND PRESERVATION OF THE MICROSCOPE 



an error of centring, which also means the error of not placing the 

 axes of the lenses in the same straight line ; so both faults are 

 described by the same term. 



It should be understood that signal green glass will not yield 

 monochromatic illumination ; only the Gifford screen or the filter 

 screen of Prof. Miethe ('/.'.) or the Nelson spectroseopic arrange- 

 ment ('/.f.) can be of real service. 



Coloured light derived from a polariscope and a selenite is not 

 monochromatic. 



For critical irork. /'<!> its test tin/ lenses or forcing out the greatest 

 resolution with the widest-angled oil-immersion lenses, daylight 

 illumination is inadmissible. 



\Vheii daylight illumination is used, a northern aspect, or at least 

 one away from direct sunlight, is to be preferred. 



It is a good plan, where it is possible, to arrange the table so that 

 the window is at the observer's left hand. The microscope should be 

 placed in a direction parallel to the window, and the light reflected 

 by the mirror through a light angle. A screen may be placed parallel 

 to the window which just allows the mirror of the microscope to 

 project beyond it. This cuts oft' direct light from the stage and 

 from the observer's eyes. 



A concave mirror with the object in its principal focus is the best 

 for diffused daylight illumination. The diaphragm should not be 

 close to the stage. When delicate microscopical work is carried on, 

 it is important to remember that the human eye can work best when 

 the body is in a state of ease. If there is any strain on the muscles 

 of the body, or if the observer is in a cramped position, vision will 

 be impaired. Consequently, where permissible, a microscope should 

 always be inclined, and the observer seated in such a way that the 

 eye can be brought to the eye-piece in a perfectly natural and com- 

 fortable manner. The body should also be steadied by resting the 

 arms on the table. 



It is advisable to use the bull's-eye as little as possible; ercn n'itJ> 

 dark-ground illumination the flat of the flame is preferable, reserving 

 the bull's-eye for those cases where the flat of the flame will not cover 

 enough of the object. Generally speaking, if the whole field is re- 

 quired to be illuminated on a dark ground, a bull's-eye will be nece> 

 sary; but for an object such as a single diatom the flat side of the 

 lamp flame will usually be large enough. 



In examining diatoms or other objects, such as the karyokinetic 

 figures in very minute nuclei of microscopic organisms, or other obscure 

 and undetermined parts of such forms of life, it is most important, 

 amongst other means, to resort to the use of large solid cones ; what 

 they teach and suggest can scarcely be neglected by the searcher for 

 the unknown. lYotes.-.or Abbedoc* not advise their employment As 

 in any way final; he says that 'the resulting image produced by 

 means of ,-i broad illuminating beam is always a mixture of a multi- 

 tude of partial image* which are more or less different and dissimilar 

 from the object it>elf;' ,-ind lie does not conceive that there is any 

 Around for expectation 'that this mixture should come nearer to a 

 .-Irictly correct projection of the object . . . than the image which 



